KANformer: Dual-priors-guided low-light enhancement via KAN and Transformer

• Published in: ACM transactions on multimedia computing communications and applications
• Main Authors: Lu, Chenyang, Wei, Zhikai, Wu, Huapeng, Sun, Le, Zhan, Tianming
• Format: Journal Article
• Language: English
• Published: 25.07.2025
• ISSN: 1551-6857; 1551-6865
• DOI: 10.1145/3750732

Images captured under low-light conditions suffer from poor visibility and clarity due to insufficient light. The emergence of deep learning has greatly boosted the development of low-light enhancement techniques and achieved promising results. However, while these low-light enhancement methods have enhanced the perceptual effects of human vision, their results in high-level visual tasks (e.g., object detection and semantic segmentation) are still unstable and even sometimes bring negative effects. Therefore, in this work, we propose a new model, KANformer, which uses a semantic-gradient prior as a guide to recover pixels relevant to the image subject from both high-frequency and low-frequency perspectives. Specifically, our model consists of three key components: Low-Frequency Enhancement module (LFE), which aims to enhance the restoration of the image subject via the semantic prior obtained from SAM; Low-Frequency-based High-frequency Enhancement module (LFHE), which utilizes the KAN module to obtain information from the low-frequency features conducive to the enhancement of high-frequency features; and Gradient-based High-frequency Enhancement module (GHE), which aims to utilize the original gradient as prior to further enhance the structural information of the image and reduce the effect of noise. In addition, we introduce the discrete wavelet transform as down-sampling method while transforming the spatial domain features to the frequency domain for processing. Experiments on multiple paired and unpaired datasets show that our method achieves better visualization and image fidelity compared to other state-of-the-art methods. In addition, experiments on object detection and segmentation show that our method provides better enhancement in improving low-light high-level vision tasks.